Refusals: BioTIERBrowse 284 families
Refusals: BioTIER
Can the model reason accurately about difficult scientific material?
Version / measure
- Domain
- Science
- Ranked models
- 5
- Reported values
- 5
- Reports
- 1 · lab
- Direction
- Contextual score
- Status
- Version not specified
Rankings
Best reported score per model across effort settings
| # | Model | Best setting | Score |
|---|---|---|---|
| 1 | Muse Spark 1.0Meta | reportedcost n/a1 value · Jul 9 · source | 98.0% |
| 2 | Muse Spark 1.1Meta | xhighcost n/a1 value · Jul 9 · source | 97.7% |
| 3 | Claude Opus 4.8Anthropic | maxcost n/a1 value · Jul 9 · source | 94.9% |
| 4 | GPT-5.5OpenAI | xhighcost n/a1 value · Jul 9 · source | 69.2% |
| 5 | Gemini 3.1 ProGoogle | highcost n/a1 value · Jul 9 · source | 57.7% |
Effort curve
Score vs. reported cost. Lines connect complete sweeps only.
No cost-linked effort sweep for this version.
Definition and comparison boundarylimited methodology
Can the model reason accurately about difficult scientific material?
Scientific knowledge and reasoning under the benchmark's published question set. The source reports a refusal percentage; interpret with the named safety objective.
Harness, tools, prompts, attempt count, and benchmark version can materially change the result. Compare within one reporting context.
Method / source ↗